Table 5. Performance measures of the SVM classifier based on input features.
Input features | Number | ACC (%) | SE (%) | SP (%) | PPV (%) | NPV (%) |
---|---|---|---|---|---|---|
MSCL, SDSCL | 10 | 54.10 | 52.63 | 56.52 | 66.67 | 41.94 |
MSCL, SDSCL, SSCL | 15 | 63.93 | 62.50 | 65.52 | 66.67 | 61.29 |
MSCL, SDSCL, NSCR | 15 | 67.21 | 63.89 | 72.00 | 76.67 | 58.06 |
MSCL, SDSCL, SSCL, NSCR | 20 | 70.49 | 70.00 | 70.97 | 70.00 | 70.97 |
MSCL, SDSCL, SSCL, MSCR, NSCR | 25 | 68.85 | 67.74 | 70.00 | 70.00 | 67.74 |
MSCL, SDSCL, SSCL, NSCR, PSD | 25 | 67.21 | 66.67 | 67.74 | 66.67 | 67.74 |
MSCL, SDSCL, SSCL, MSCR, NSCR, PSD | 30 | 67.21 | 67.86 | 66.67 | 63.33 | 70.97 |
Abbreviations: ACC, accuracy; SE, sensitivity; SP, specificity.